Abstract: Configural and featural processing, two most important processing modes in face recognition, were reviewed based on literatures frequently cited and supported during recent 40 years. Different related conceptions and methods were discussed, and the findings supported both by behavioral and cognitive neuroscience researches were emphasized and systematically interpreted under a frame of dissociated facial recognition. Developmental and neurobiological studies were further reviewed to provide a robust evidence of two processing modes in facial recognition. Furthermore, the potential different levels of configural processing were proposed to explain the conflicted conclusions in the field. Whatever, the distinction between configural processing and featural processing was well supported, and these two processing pattern may be used to account for how we obtain different facial property information from one’s faces. Still, the subcategory of configural information and the differences between conceptions such as configural, holistic, whole, and global were suggested to be explored in the future.

Key words: face recognition, configural processing, featural processing